与当前最优的从零开始训练的模型相比，Moirai 的零样本（zero-shot）预测能力具有竞争力，甚至在某些情况下表现更优。此外，还发布并开源了预训练框架 uni2ts、Moirai 模型权重，以及 LOTSA 数据集——这是目前最大的开源时间序列预测预训练数据集。论文题目: 《Unified Training of Universal Time Series Forecasting...
数据分布和任务分布的设计是预训练流程中的两个关键方面。这种设计赋予了“大规模时间序列模型”(Large Time Series Model, LTM)多样的能力,使其能够适应各种下游任务。这种灵活性与当前深度预测范式形成对比,在深度预测范式中,模型通常针对特定的数据集和设置进行专业化。 Moriai的效果 Moriai训练了三种规模的 Moirai ...
Voice2Series(Yang等,2021)利用预训练语音处理模型的表示学习能力,将语音数据作为单变量时间信号用于时间序列分类,这是第一个能够为时间序列任务提供重新编程的框架。 在Voice2Series之后,出现了几个基于对比学习的技术的时间序列数据的预训练基础模型,包括TF-C、TS2Vec和CLUDA。TF-C包含基于时间的组件和基于频率的...
We release a open codebase OpenLTM to explore the design philosophy of large time-series models, which contains a simple pipeline to train large time-series models :) Timer (Large Time-Series Model) This repo provides official code, datasets and checkpoints for Timer: Generative Pre-trained Tr...
TimeMixer++: A General Time Series Pattern Machine for Universal Predictive Analysis Root Cause Analysis of Anomalies in Multivariate Time Series through Granger Causal Discovery Amortized Control of Continuous State Space Feynman-Kac Model for Irregular Time Series ...
Official code, datasets and checkpoints for "Timer: Generative Pre-trained Transformers Are Large Time Series Models" (ICML 2024) - Large-Time-Series-Model/scripts/forecast/README.md at main · thuml/Large-Time-Series-Model
One or more modeling tasks are executed in parallel on the computing platform by, for each modeling task, training a model using all the time series in the group of time series of the corresponding modeling task.QUANZ, BRIAN LEOGIFFORD, WESLEY, M....
In this paper we propose a new method for time series pattern classification. It is based on the generative modeling using Autoregressive(AR) model and optimizing the boundaries between these models using the large margin concepts. The developed model captures the correlations in the time series da...
A Time Series Model is defined as a method used in various scientific fields to analyze data exhibiting patterns like trends, seasonal fluctuations, and irregular cycles. It involves building linear models, such as ARMA and ARIMA models, to forecast future observations, estimate the impact of inter...
原始题目:Position: What Can Large Language Models Tell Us about Time Series Analysis 中文翻译:立场:关于时间序列分析,大型语言模型能告诉我们什么 发表时间:2024-06-01 平台:ICML 文章链接:http://arxiv.org/abs/2402.02713 开源代码:NA 摘要时间序列分析对于理解各种现实世界系统和应用程序中固有的复杂性至关...